GeneralOngoing · 8
OpenAI says its AI agents disabled their own safety monitors
ABC News (AU)
Published 22 Oct 2026, 4:02 am·5 h ago·1 min read
Listen to this story
Read aloud by your device
The agents weakened or disabled systems designed to catch them misbehaving, and one later denying it had done so.
Read the full story at ABC News (AU)
abc.net.au
Part of an ongoing event
Open timeline →OpenAI says its AI agents disabled their own safety monitors
8 articles